Davy, Christina M.

Evaluating the impact of climate change is arguably one of the main goals of conservation biology, which can be addressed in part by studying the demographic history of species in the region of interest. In North America, landscape and species composition during the most recent Pleistocene epoch was primarily influenced by glaciation cycles. Glacial advance and retreat caused species ranges to shift as well, leaving signatures of past population bottlenecks in the genetic code of most species. Genomic tools have shown to be important tools for understanding these demographic events to enhance conservation biology measures in several species. In my thesis I first reviewed the state of ungulate genomics, with a focus on how such data sets can be used in understand demography, adaptation, and inform conservation and management. Importantly, the review introduces key analyses like the pairwise sequentially Markovian coalescent and features like variation in antlers and horns and selection pressures that are used throughout subsequent chapters. Using the North American mountain goat as a model species, I then explored the genomic and phenotypic variation in this alpine specialist mammal. Starting with the generation of the first genome assembly for the mountain goat, I identified genes unique to the mountain goat and modeled demographic history going back millions of years using a pairwise sequentially Markovian coalescent approach. Species' effective population size generally paralleled climatic trends over the past one hundred thousand years and severely declined to under a thousand individuals during the last glacial maximum. Given the biological importance of horns in mountain goats and the recent scientific interest in genetic basis of headgear, I analyzed over 23,000 horn records from goats harvested in British Columbia, Alaska and Northwest Territories from 1980 to 2017. Overall, variation in horn size over space and time was low; goats harvested further North had shorter horn lengths and smaller horn circumferences in one year old and 4 years and older age classes and 4 years and older age class, respectively. Proximity of roads, which was used as an indicator of artificial selection, had a small effect on horn size, with larger horns being harvested closer to major roads. Finally, I used two range-wide genomic data sets sequenced with a whole genome re-sequencing and reduced representation approaches to provide estimates of genetic diversity, contemporary effective population sizes and population structure. These insights can help inform management and will potentially make an impact in preserving the mountain goat.

Author Keywords: genome assembly, horn size, Oreamnos americanus, population demography, reduced representation sequencing, whole genome resequencing